Date: Tue, 5 Jan 2016 13:42:49 -0700 From: Warner Losh <imp@bsdimp.com> To: Karl Denninger <karl@denninger.net> Cc: "freebsd-arm@freebsd.org" <freebsd-arm@freebsd.org> Subject: Re: How did I manage to break this?! Message-ID: <CANCZdfq4V6H=ZBZ83S3OvsJCCXiojs81vcvrgO%2BFKYu-SS7Rdw@mail.gmail.com> In-Reply-To: <568C2A57.2070406@denninger.net> References: <568C1919.2090407@denninger.net> <CANCZdfq1i-napivz=2PxLz7yCkbP8vfDhQvjjAD=NiOk0dMQxQ@mail.gmail.com> <568C1B27.3050206@denninger.net> <CANCZdfqWbtXHz=SdyNJH3RWa-J30FuYMCSMmdJXJ_5tW9xUB_Q@mail.gmail.com> <568C2258.1040605@denninger.net> <CANCZdfo3cUFqUpuAMdeCq9kv7EEg9Tq68hY=pEBryiBZEGpRsw@mail.gmail.com> <568C2A57.2070406@denninger.net>
next in thread | previous in thread | raw e-mail | index | archive | help
I'm reasonably sure you are right: armv6hf isn't involved for either one of us. It may be the pkg repo. There's a small chance that something I did to the linker may have caused this. Lemme build an image w/o the recent change to see if that's at fault. Warner On Tue, Jan 5, 2016 at 1:40 PM, Karl Denninger <karl@denninger.net> wrote: > I'm reasonably sure that armv6hf isn't involved; I canvassed the system > that was updated and unless I missed something really ridiculous there's > no hf-code left on the box. > > This looks like something got broken in the pkg repo that is being > picked up by the pkg utility... I can of course build the ports myself > but was hoping to not need to. > > On 1/5/2016 14:39, Warner Losh wrote: > > On a newly built nanobsd image that I hacked to allow enough space to > > install packages, > > I've recreated this problem with armv6hf being in the loop at all. > > > > Warner > > > > On Tue, Jan 5, 2016 at 1:06 PM, Karl Denninger <karl@denninger.net > > <mailto:karl@denninger.net>> wrote: > > > > $ readelf -h git > > ELF Header: > > Magic: 7f 45 4c 46 01 01 01 00 00 00 00 00 00 00 00 00 > > Class: ELF32 > > Data: 2's complement, little endian > > Version: 1 (current) > > OS/ABI: NONE > > ABI Version: 0 > > Type: EXEC (Executable file) > > Machine: ARM > > Version: 0x1 > > Entry point address: 0xbe08 > > Start of program headers: 52 (bytes into file) > > Start of section headers: 1623892 (bytes into file) > > Flags: 0x5000202, Version5 EABI, has > > entry > > point, software FP > > Size of this header: 52 (bytes) > > Size of program headers: 32 (bytes) > > Number of program headers: 8 > > Size of section headers: 40 (bytes) > > Number of section headers: 28 > > Section header string table index: 27 > > > > And the allegedly-not-there shared library: > > > > ELF Header: > > Magic: 7f 45 4c 46 01 01 01 00 00 00 00 00 00 00 00 00 > > Class: ELF32 > > Data: 2's complement, little endian > > Version: 1 (current) > > OS/ABI: NONE > > ABI Version: 0 > > Type: DYN (Shared object file) > > Machine: ARM > > Version: 0x1 > > Entry point address: 0x1830 > > Start of program headers: 52 (bytes into file) > > Start of section headers: 37324 (bytes into file) > > Flags: 0x5000202, Version5 EABI, has > > entry > > point, software FP > > Size of this header: 52 (bytes) > > Size of program headers: 32 (bytes) > > Number of program headers: 5 > > Size of section headers: 40 (bytes) > > Number of section headers: 29 > > Section header string table index: 26 > > > > > > > > On 1/5/2016 14:00, Warner Losh wrote: > > > what dpes readelf -h on all the affected files tell you? You're > > most > > > interested in the line: > > > > > > Flags: 0x5000202, Version5 EABI, has > > > entry point, software FP > > > > > > > > > Warner > > > > > > On Tue, Jan 5, 2016 at 12:36 PM, Karl Denninger > > <karl@denninger.net <mailto:karl@denninger.net> > > > <mailto:karl@denninger.net <mailto:karl@denninger.net>>> wrote: > > > > > > $ uname -v > > > FreeBSD 11.0-CURRENT #0 r293189: Tue Jan 5 08:44:05 CST 2016 > > > > > karl@NewFS.denninger.net: > /pics/CrossBuild/obj/arm.armv6/pics/CrossBuild/src/sys/RPI2 > > > > > > Built a few hours ago, svn'd this morning. rm -r'd both the > > > object and > > > export directories before starting to prevent the risk of a > > > carry-over. > > > > > > And no, I pkg delete'd all the packages on the machine > > (including pkg > > > itself!) and then rm -r'd the entire /usr/local/lib directory, > > > then used > > > pkg to reload them. > > > > > > No chance they're leftovers. > > > > > > On 1/5/2016 13:33, Warner Losh wrote: > > > > what revision are you at? > > > > > > > > Could these be left-over hard-float libraries? > > > > > > > > Warner > > > > > > > > On Tue, Jan 5, 2016 at 12:27 PM, Karl Denninger > > > <karl@denninger.net <mailto:karl@denninger.net> > > <mailto:karl@denninger.net <mailto:karl@denninger.net>> > > > > <mailto:karl@denninger.net <mailto:karl@denninger.net> > > <mailto:karl@denninger.net <mailto:karl@denninger.net>>>> wrote: > > > > > > > > Went back from arm6hf to arm6 on a Pi2 (cross-built the > > > world and > > > > kernel, then rsync'd it), and now I'm getting this > > with two > > > packages > > > > that I loaded from the "pkg add" command (which, > > > incidentally, was the > > > > reason to do it in the first place, since arm6hf has > > no package > > > > repository) > > > > > > > > The system itself is running fine. > > > > > > > > $ git > > > > Shared object "libintl.so.8" not found, required by "git" > > > > > > > > Bee-ess > > > > > > > > $ ldconfig -r|grep intl > > > > 134:-lintl.8 => /usr/local/lib/libintl.so.8 > > > > 135:-lintl.9 => /usr/local/lib/libintl.so.9 > > > > > > > > $ ls -al /usr/local/lib/*intl* > > > > -rw-r--r-- 1 root wheel 74122 Dec 7 12:58 > > > /usr/local/lib/libintl.a > > > > lrwxr-xr-x 1 root wheel 16 Dec 7 12:58 > > > > /usr/local/lib/libintl.so > > > > -> libintl.so.8.1.4 > > > > lrwxr-xr-x 1 root wheel 16 Dec 7 12:58 > > > > /usr/local/lib/libintl.so.8 -> libintl.so.8.1.4 > > > > -rw-r--r-- 1 root wheel 51832 Dec 7 12:58 > > > > /usr/local/lib/libintl.so.8.1.4 > > > > lrwxr-xr-x 1 root wheel 12 Dec 7 12:58 > > > > /usr/local/lib/libintl.so.9 -> libintl.so.8 > > > > > > > > That all came in from the pkg add, so why is the loader > > > > complaining that > > > > libintl.so.8 not there? Incidentally, bash throws up > > on the > > > same > > > > error. > > > > > > > > -- > > > > Karl Denninger > > > > karl@denninger.net <mailto:karl@denninger.net> > > <mailto:karl@denninger.net <mailto:karl@denninger.net>> > > > <mailto:karl@denninger.net <mailto:karl@denninger.net> > > <mailto:karl@denninger.net <mailto:karl@denninger.net>>> > > > > <mailto:karl@denninger.net <mailto:karl@denninger.net> > > <mailto:karl@denninger.net <mailto:karl@denninger.net>> > > > <mailto:karl@denninger.net <mailto:karl@denninger.net> > > <mailto:karl@denninger.net <mailto:karl@denninger.net>>>> > > > > /The Market Ticker/ > > > > /[S/MIME encrypted email preferred]/ > > > > > > > > > > > > > > -- > > > Karl Denninger > > > karl@denninger.net <mailto:karl@denninger.net> > > <mailto:karl@denninger.net <mailto:karl@denninger.net>> > > > <mailto:karl@denninger.net <mailto:karl@denninger.net> > > <mailto:karl@denninger.net <mailto:karl@denninger.net>>> > > > /The Market Ticker/ > > > /[S/MIME encrypted email preferred]/ > > > > > > > > > > -- > > Karl Denninger > > karl@denninger.net <mailto:karl@denninger.net> > > <mailto:karl@denninger.net <mailto:karl@denninger.net>> > > /The Market Ticker/ > > /[S/MIME encrypted email preferred]/ > > > > > > -- > Karl Denninger > karl@denninger.net <mailto:karl@denninger.net> > /The Market Ticker/ > /[S/MIME encrypted email preferred]/ >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CANCZdfq4V6H=ZBZ83S3OvsJCCXiojs81vcvrgO%2BFKYu-SS7Rdw>